Quantization made by Richard Erkhov.

MN-12B-Starsong-v1 - GGUF

Model creator: https://huggingface.co/aetherwiing/
Original model: https://huggingface.co/aetherwiing/MN-12B-Starsong-v1/

Name	Quant method	Size
MN-12B-Starsong-v1.Q2_K.gguf	Q2_K	4.46GB
MN-12B-Starsong-v1.IQ3_XS.gguf	IQ3_XS	4.94GB
MN-12B-Starsong-v1.IQ3_S.gguf	IQ3_S	5.18GB
MN-12B-Starsong-v1.Q3_K_S.gguf	Q3_K_S	5.15GB
MN-12B-Starsong-v1.IQ3_M.gguf	IQ3_M	5.33GB
MN-12B-Starsong-v1.Q3_K.gguf	Q3_K	5.67GB
MN-12B-Starsong-v1.Q3_K_M.gguf	Q3_K_M	5.67GB
MN-12B-Starsong-v1.Q3_K_L.gguf	Q3_K_L	6.11GB
MN-12B-Starsong-v1.IQ4_XS.gguf	IQ4_XS	6.33GB
MN-12B-Starsong-v1.Q4_0.gguf	Q4_0	6.59GB
MN-12B-Starsong-v1.IQ4_NL.gguf	IQ4_NL	6.65GB
MN-12B-Starsong-v1.Q4_K_S.gguf	Q4_K_S	6.63GB
MN-12B-Starsong-v1.Q4_K.gguf	Q4_K	6.96GB
MN-12B-Starsong-v1.Q4_K_M.gguf	Q4_K_M	6.96GB
MN-12B-Starsong-v1.Q4_1.gguf	Q4_1	7.26GB
MN-12B-Starsong-v1.Q5_0.gguf	Q5_0	7.93GB
MN-12B-Starsong-v1.Q5_K_S.gguf	Q5_K_S	7.93GB
MN-12B-Starsong-v1.Q5_K.gguf	Q5_K	8.13GB
MN-12B-Starsong-v1.Q5_K_M.gguf	Q5_K_M	8.13GB
MN-12B-Starsong-v1.Q5_1.gguf	Q5_1	8.61GB
MN-12B-Starsong-v1.Q6_K.gguf	Q6_K	9.37GB
MN-12B-Starsong-v1.Q8_0.gguf	Q8_0	12.13GB

Original model description:

base_model: - nothingiisreal/MN-12B-Celeste-V1.9 - Sao10K/MN-12B-Lyra-v1 library_name: transformers tags: - mergekit - merge

Mistral Nemo 12B Starsong

This is a merge of pre-trained language models created using mergekit. Just messing around, thought this one turned out distinct enough? YMMV on that, but I kinda liked it. It's definitely better for SFW than for NSFW tho. Seems to be stabler with Mistral formatting? Weird, both merged models were trained with ChatML. I guess we chalk it up to another merging anomaly, heh.
Unlicensed bc I got enlightened (lmao) and don't want to license machine output anymore.
Thanks to Sao for Lyra btw, really liking the direction where those experimental models are heading. Keep it up!

Static GGUF (by Mradermacher)
EXL2 (by kingbri of RoyalLab)

Merge Details

Merge Method

This model was merged using the TIES merge method using nothingiisreal/MN-12B-Celeste-V1.9 as a base.

Merge Fodder

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Sao10K/MN-12B-Lyra-v1
    parameters:
      density: 0.45
      weight: 0.5
  - model: nothingiisreal/MN-12B-Celeste-V1.9
    parameters:
      density: 0.65
      weight: 0.5

merge_method: ties
base_model: nothingiisreal/MN-12B-Celeste-V1.9
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16